BioMedInformatics,
Journal Year:
2024,
Volume and Issue:
4(1), P. 837 - 852
Published: March 14, 2024
This
review
explores
the
transformative
integration
of
artificial
intelligence
(AI)
and
healthcare
through
conversational
AI
leveraging
Natural
Language
Processing
(NLP).
Focusing
on
Large
Models
(LLMs),
this
paper
navigates
various
sections,
commencing
with
an
overview
AI’s
significance
in
role
AI.
It
delves
into
fundamental
NLP
techniques,
emphasizing
their
facilitation
seamless
conversations.
Examining
evolution
LLMs
within
frameworks,
discusses
key
models
used
healthcare,
exploring
advantages
implementation
challenges.
Practical
applications
conversations,
from
patient-centric
utilities
like
diagnosis
treatment
suggestions
to
provider
support
systems,
are
detailed.
Ethical
legal
considerations,
including
patient
privacy,
ethical
implications,
regulatory
compliance,
addressed.
The
concludes
by
spotlighting
current
challenges,
envisaging
future
trends,
highlighting
potential
reshaping
interactions.
Scientific Reports,
Journal Year:
2024,
Volume and Issue:
14(1)
Published: April 11, 2024
Abstract
Free-text
analysis
using
machine
learning
(ML)-based
natural
language
processing
(NLP)
shows
promise
for
diagnosing
psychiatric
conditions.
Chat
Generative
Pre-trained
Transformer
(ChatGPT)
has
demonstrated
preliminary
initial
feasibility
this
purpose;
however,
whether
it
can
accurately
assess
mental
illness
remains
to
be
determined.
This
study
evaluates
the
effectiveness
of
ChatGPT
and
text-embedding-ada-002
(ADA)
model
in
detecting
post-traumatic
stress
disorder
following
childbirth
(CB-PTSD),
a
maternal
postpartum
affecting
millions
women
annually,
with
no
standard
screening
protocol.
Using
sample
1295
who
gave
birth
last
six
months
were
18+
years
old,
recruited
through
hospital
announcements,
social
media,
professional
organizations,
we
explore
ChatGPT’s
ADA’s
potential
screen
CB-PTSD
by
analyzing
narratives.
The
PTSD
Checklist
DSM-5
(PCL-5;
cutoff
31)
was
used
CB-PTSD.
By
developing
an
ML
that
utilizes
numerical
vector
representation
ADA
model,
identify
via
narrative
classification.
Our
outperformed
(F1
score:
0.81)
previously
published
large
text-embedding
models
trained
on
health
or
clinical
domains
data,
suggesting
harnessed
modeling
approach
could
generalized
other
disorders.
Research Square (Research Square),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Nov. 2, 2023
Abstract
In
the
realm
of
social
media,
users
frequently
convey
personal
sentiments,
with
some
potentially
indicating
cognitive
distortions
or
suicidal
tendencies.
Timely
recognition
such
signs
is
pivotal
for
effective
interventions.
response,
we
introduce
two
novel
annotated
datasets
from
Chinese
focused
on
and
risk
classification.
We
propose
a
comprehensive
benchmark
using
both
supervised
learning
large
language
models,
especially
GPT
series,
to
evaluate
performance
these
datasets.
To
assess
capabilities
employed
three
strategies:
zero-shot,
few-shot,
fine-tuning.
Furthermore,
deeply
explored
analyzed
models
psychological
perspective,
shedding
light
their
strengths
limitations
in
identifying
understanding
complex
human
emotions.
Our
evaluations
underscore
difference
between
approaches,
often
challenged
by
subtle
category
distinctions.
While
GPT-4
consistently
delivered
strong
results,
GPT-3.5
showed
marked
improvement
suicide
classification
after
This
research
groundbreaking
its
evaluation
media
tasks,
accentuating
models'
potential
contexts.
All
code
are
made
available
at:
\url{https://github.com/HongzhiQ/SupervisedVsLLM-EfficacyEval}.
Transactions on Social Science Education and Humanities Research,
Journal Year:
2024,
Volume and Issue:
4, P. 99 - 109
Published: March 12, 2024
With
the
development
of
artificial
intelligence,
ChatGPT
has
set
off
a
hot
wave
in
various
fields,
and
brings
opportunities
challenges
to
education.
The
positive
negative
effects
on
education
have
aroused
heated
discussion
academic
circles.
Based
its
principle
process,
continuous
optimization
language
model,
can
play
unique
advantage
oral
English
teaching,
such
as:
providing
personalized
practice,
real-time
error
correction
feedback,
real
dialogue
simulation,
safe
comfortable
environment.
At
same
time,
challenge
teaching
is
also
direction
future
technological
improvement,
communication
spoken
language,
recognition
voice
tone
intonation,
integration
with
VR
technology,
autonomous
assessment
level,
etc.,
Those
become
application
prospects
teaching.
Artificial
intelligence
assisted
new
mode
The
recent
public
release
of
the
generative
AI
language
model
ChatGPT
has
captured
imagination
and
resulted
a
rapid
uptake
widespread
experimentation
by
general
academia
alike.
number
academic
publications
focusing
on
capabilities
as
well
practical
ethical
implications
been
growing
exponentially.
One
concerns
with
this
unprecedented
growth
in
scholarship
related
to
AI,
particular
ChatGPT,
is
that
most
cases
raw
data,
text
original
‘conversations,’
have
not
made
available
audience
papers
thus
cannot
be
drawn
assess
veracity
arguments
conclusions
therefrom.
This
paper
provides
protocol
for
documentation
archiving
these
data.
Research Square (Research Square),
Journal Year:
2024,
Volume and Issue:
unknown
Published: Feb. 26, 2024
Abstract
Free-text
analysis
using
Machine
Learning
(ML)-based
Natural
Language
Processing
(NLP)
shows
promise
for
diagnosing
psychiatric
conditions.
Chat
Generative
Pre-trained
Transformer
(ChatGPT)
has
demonstrated
preliminary
initial
feasibility
this
purpose;
however,
whether
it
can
accurately
assess
mental
illness
remains
to
be
determined.
This
study
evaluates
the
effectiveness
of
ChatGPT
and
text-embedding-ada-002
(ADA)
model
in
detecting
post-traumatic
stress
disorder
following
childbirth
(CB-PTSD),
a
maternal
postpartum
affecting
millions
women
annually,
with
no
standard
screening
protocol.
Using
sample
1,295
who
gave
birth
last
six
months
were
18
+
years
old,
recruited
through
hospital
announcements,
social
media,
professional
organizations,
we
explore
ChatGPT’s
ADA’s
potential
screen
CB-PTSD
by
analyzing
narratives
only.
The
PTSD
Checklist
DSM-5
(PCL-5;
cutoff
31)
was
used
CB-PTSD.
By
developing
an
ML
that
utilizes
numerical
vector
representation
ADA
model,
identify
via
narrative
classification.
Our
outperformed
(F1
score:
0.82)
previously
published
large
language
models
(LLMs)
trained
on
health
or
clinical
domains
data,
suggesting
harnessed
modeling
approach
could
generalized
other
disorders.
1
